Nonviral receptor-mediated gene transfer systems with improved nuclear import characteristics

Research project P 14289 DNA complexes for improved nuclear import Ernst WAGNER 06.03.2000 Nonviral receptor-mediated gene transfer systems with improved nuclear import characteristics Increasing numbers of clinical protocols utilize nonviral gene transfer systems. However, the lack of effective vectors is a major bottleneck for progress in somatic gene therapy. As previously demonstrated by several research groups including our lab, one major barrier for nonviral gene transfer particles (in contrast to e.g. recombinant adenovirus) is the import into the nucleus of cells. While dividing cells could incorporate the exogenous DNA passively into the nucleus during mitosis, specific nuclear import mechanisms are considered to be necessary in slowly dividing or non-dividing cells. Motifs of nuclear import peptide signals have been described that are recognized either directly by a transport receptor or indirectly via special adapters (importins and transportins) mediating interaction with the nuclear pore complex and triggering translocation through the pore into the nucleus. The aim of our studies is the design and characterization of novel gene transfer systems with improved efficiency for nuclear import. These gene transfer systems are based on DNA/ligand-polycation complexes with cell-binding ligands for receptor-mediated uptake and functions for enhanced endosomal release and protection of the DNA against degradation. Nuclear import elements will be incorporated by i) providing the DNA with a nucleic acid cassette able to bind to host cell factors that are subject to the nuclear import machinery ii) covalently coupling an import signal to the DNA or iii) binding an import signal to the DNA by adapter molecules such as peptide nucleic acids, polycations, or intercalating agents. The new gene transfer systems will be tested in vitro (including evaluation in normal cells and tumor cells, and in a centrifugal elutriation protocol) and in vivo in two murine models (systemic delivery of DNA complexes into tumors; and DNA delivery into the skin of mice).

Nonviral gene vectors can efficiently transfer their gene into dividing (mitotic) cells only. A main reason is the inefficiency of crossing the cell nuclear membrane barrier which is prerequisite for transcription and expression. This characteristic presents a real practical problem, as most body cells are non-dividing or slowly dividing, even in pathological cases such as with tumor cells. Aim of the project was the improval of nuclear import for nonviral vectors. For attachment of nuclear localisation (NLS) elements into DNA complexes, we (a) covalently coupled an import signal directly to the DNA or (b) complexed the DNA with an import signal coupled to a polycation, or (c) investigated the influence of the SV40 enhancer/early promoter region on the cell cycle dependence of nonviral gene transfer. The new formulations containing DNA, NLS elements, and the conjugates of transferrin with polycations were tested in standard in transfections protocols. However, incorporation of NLS elements did not show any positive influence on gene transfer of cells in non-mitotic phases of the cell cycle. In contrast, electrical mediated gene transfer (electroporation) was able to transfect nondividing cells and cells in the G1 phase . In addition, special DNA polyplexes containing the linear (but not branched form) form of the polycation polyethylenimine also were able to transfect nondividing cells. In collaboration with M. Ogris (LMU Munich) it was found that incorporation of a membrane-active peptide derived from bee venom melittin also was able to transfer DNA polyplexes out of endosomal vesicles into the nucleus of cells. The gene transfer technology was applied in collaboration with J. Seipelt and E. Küchler for the delivery of novel therapeutic gene expression constructs into human tumor cells. Using bicistronic expression constructs, the cytotoxic activity of human rhinovirus protease 2A was combined with an IRES controlled expression cassette for human interleukin-2. Protease 2A blocks the standard cellular host protein synthesis, but not IRES-controlled protein synthesis. Consistently, high expression of immunomodulatory human interleukin-2 was found despite shut-off of host protein synthesis. This strategy opens a new concept for future cancer gene therapies.